Review of Voice Stress Based Technologies For the Detection of Deception

Voice-Based Lie Detectors

For several years police departments and federal agencies have been the target of determined marketing efforts by the makers of voice-based lie detectors. The principal sales pitch since the early 1970s has been that their particular voice device is more accurate and cost less than the polygraph (or any of the competing voice lie detectors). These sales pitches appeal to cash-strapped departments and agencies, and if true, the devices would quickly come to replace the current gold standard, the polygraph. There are at least 10 different voice devices sold in the US, freeware packages are on the Internet, and a burgeoning list of smartphone applications. Why, then, have voice devices not come to surpass the polygraph in law enforcement and government over the past 40 years?

It is a fact that voice lie detectors trail far behind the polygraph, even in sheer number of units. And though some voice device manufacturers claim to have sold X number of their devices to various organizations, a significant number of departments discarded them quickly after discovering that the reality of the systems fall far short of what manufacturers had promised.

Voice device manufacturers would blame their limited market impact on the bias of polygraph examiners, but the truth goes deeper. The first reason for the failure of voice lie detectors to capture the field is simple: they are highly unreliable. More than that, they have been discredited by a large and expanding body of independent scientific research. This is the main reason they are banned by the US Department of Defense. With such dismal accuracy they are little more than very expensive interrogation props. See the table below for the published non-manufacturer’s research on voice-based systems.

Summary of Published Voice Lie Detector Research

VSA Research Table

Perhaps the US National Academy of Sciences (2003) said it best:

Overall, this research and the few controlled tests conducted over the past decade offer little or no scientific basis for the use of the computer voice stress analyzer or similar voice measurement instruments as an alternative to the polygraph for the detection of deception. The practical performance of voice stress analysis for detecting deception has not been impressive.

A second reason the sales of voice devices have been limited is that diligent departments have learned there really is no cost savings over the polygraph in terms of instrumentation and training. Indeed, their bottom lines are almost the same. Claims otherwise are simply false, as the following comparison shows:

Polygraph

Instrument: $  6000

Tuition: $  5000

Total: $11,000

Voice Stress

CVSATM Program: $  9950

Tuition: $  1340

Total: $11,290

Despite miraculous claims for what voice devices can do, there are two things they can’t: detect deception accurately and save tax dollars. This does not even consider the hidden cost of litigation some departments have recently experienced due to errors caused by voice stress devices.

When the facts are known, it’s clear why the polygraph is still the leading technology for deception detection. Again, to quote the National Academy of Sciences: [S]ome potential alternatives to the polygraph show promise, but none has yet been shown to outperform the polygraph.

If accuracy and cost effectiveness are important, the polygraph is the right choice.

AttachmentSize
vsa_research_table.jpg353.97 KB